INDEX

Explanations

seeking help

The neuron fires on promotional or marketing-style tokens—brand names, services, and calls-to-action.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 אך

0.83

 لكن

0.83

 pertanto

0.81

したがって

0.80

BUT

0.77

但也

0.76

従って

0.75

 किंतु

0.73

umably

0.73

 परन्तु

0.73

POSITIVE LOGITS

1.61

！

1.33

 today

1.31

 yourself

1.31

!</

1.28

!»

1.27

!”

1.20

!`

1.17

!’

1.16

!“

1.16

Activations Density 0.800%